On Formalizing Fairness in Prediction with Machine Learning

TL;DR

This paper surveys various formalizations of fairness in machine learning prediction algorithms, analyzing their theoretical and empirical strengths and limitations, and proposes new notions inspired by social justice theories.

Contribution

It provides a comprehensive review of fairness formalizations, critiques their applicability, and introduces new fairness notions based on distributive justice.

Findings

Certain fairness formalizations are limited in domain applicability.

Empirical critiques reveal gaps in existing fairness measures.

Proposed fairness notions address specific social justice critiques.

Abstract

Machine learning algorithms for prediction are increasingly being used in critical decisions affecting human lives. Various fairness formalizations, with no firm consensus yet, are employed to prevent such algorithms from systematically discriminating against people based on certain attributes protected by law. The aim of this article is to survey how fairness is formalized in the machine learning literature for the task of prediction and present these formalizations with their corresponding notions of distributive justice from the social sciences literature. We provide theoretical as well as empirical critiques of these notions from the social sciences literature and explain how these critiques limit the suitability of the corresponding fairness formalizations to certain domains. We also suggest two notions of distributive justice which address some of these critiques and discuss avenues for prospective fairness formalizations.